SSRAM

A Segmentation-based Sequence Residual Attention Model for KRAS Gene Mutation Status Prediction in Colorectal Cancer

1.introduction

In this paper, we propose a Segmentation-based Sequence Residual Attention Model (SSRAM) to predict patients' KRAS gene mutation status with the segmentation task facilitating the classification task.

2.Requirement

PyTorch='1.7.1'